| 05:36:49 | jacques | ARG |
| 05:36:57 | jacques | fscking april 1 |
| 05:36:58 | Russ | yes? |
| 05:37:17 | Russ | not for another hour and 19 minutes |
| 05:37:26 | jacques | well freshmeat has already redone their site |
| 05:37:35 | jacques | they are going to devite it to the xbox apparently |
| 05:37:44 | jacques | This is the year of the Xbox, which we at freshmeat consider the most important development in the history of console gaming. |
| 05:37:49 | jacques | etc etc |
| 05:38:02 | TomW | had you going, didn't they? |
| 05:38:16 | jacques | TomW: yes for a second or two they did |
| 05:38:42 | TomW | LOL that was before reality crept in? |
| 05:38:50 | jacques | heh, they are going to be adopting passport technology |
| 05:39:03 | jacques | yes I had forgotten april 1 |
| 05:39:31 | jacques | i guess i should go look at other popular sites to see what they've done |
| 05:39:39 | TomW | Maybe you should go to the Micro$oft site and see if they have a positive article about Linux? |
| 05:39:59 | jacques | LOL |
| 05:40:13 | jacques | heh the comments on the fm article sound like they dont get the joke |
| 05:40:17 | jacques | that's even funnier |
| 05:40:40 | jacques | one guy's trying to convince them to not use passport |
| 05:40:43 | TomW | must be burned out programmers.. |
| 05:41:05 | jacques | heh |
| 05:41:29 | Russ | heh |
| 05:41:36 | Russ | did you notice the hyperlink for /.? |
| 05:41:42 | Russ | go ahead, click |
| 05:42:35 | jacques | gtta try that |
| 05:43:12 | jacques | hmm where? |
| 05:43:58 | Russ | on the freshmeat.NET article |
| 05:44:47 | jacques | ah |
| 05:45:33 | jacques | aaah that's gross |
| 05:45:40 | jacques | i cant believe they did that |
| 05:45:44 | TomW | ah, krap, that's gross! |
| 05:46:30 | TomW | I guess that tears it on the issue of subscriptions? |
| 05:46:56 | jacques | im gonna have nightmares now |
| 05:47:42 | TomW | The "really cool site" is a link to ZiffDavis... |
| 05:47:44 | jacques | hi andersee |
| 05:47:50 | jacques | heh |
| 05:48:00 | jacques | im afraid to click on things now |
| 05:48:45 | TomW | "... the Passport RFC ..." |
| 05:49:08 | Russ | you didn't know about goat sex? |
| 05:49:36 | Russ | and *never* to click a link that says goat sex? |
| 05:51:05 | jacques | well i had heard that but I wasn't looking at where the link was going |
| 05:52:40 | jacques | im trying to get my tinyx tslib driver working |
| 05:52:57 | Russ | that'd be cool |
| 05:53:02 | Russ | I want a xdmcp image |
| 05:54:38 | jacques | i can get it to all compile and link, just not actually work |
| 05:54:50 | jacques | i have the tslib test stuff working so it should only be a matter of time |
| 05:55:02 | jacques | longest part is transferring the x server to the tuxscreen |
| 05:55:24 | TomW | jacques: I did have 16Meg running on the mainboard, then I toasted the board before I could get blob to take it to 32Meg. |
| 05:55:56 | jacques | TomW: yeah I read the backlog and was pretty excited to see that |
| 05:56:08 | jacques | TomW: sorry about the toast tho :-\ |
| 05:56:27 | jacques | i would pay to have mine upgraded |
| 05:57:03 | TomW | jacques: I have another baord that the use of an SODIMM will crash the board, that one is getting the larger DRAM. I promise I won't toast it this time. :-( |
| 05:57:22 | jacques | :-D |
| 05:57:34 | jacques | all the tuxscreen needs is more ram |
| 05:58:07 | TomW | Doesn't look like we can expand the mainboard > 32M without a lot of pain. |
| 05:58:23 | jacques | that's soo much better than 8MB |
| 05:58:59 | TomW | I was wondering if it would be possible to use one of the Tuxes as a server... Keep the script kiddies guessing. |
| 05:59:34 | Russ | TomW: get dtk if you like to keep them guessing |
| 05:59:35 | jacques | heh |
| 05:59:35 | Russ | port scan me |
| 05:59:59 | TomW | IP? |
| 06:00:29 | TomW | ping russ |
| 06:00:42 | Russ | /whois |
| 06:00:46 | Russ | or just use russ.dhs.org |
| 06:01:02 | andersee | Russ: tried semaphores? |
| 06:01:27 | Russ | andersee: not yet |
| 06:01:37 | Russ | going to try to build dnrd |
| 06:01:45 | jacques | hmm looks like I mangled it so badly it's not even trying to init the ts anymore :-) |
| 06:01:54 | andersee | k. Lemme know if you have problems... |
| 06:01:55 | jacques | should be easy enough to fix |
| 06:03:03 | TomW | Russ: nmap doesn't see anything. |
| 06:03:39 | Russ | ? |
| 06:03:44 | Russ | oh |
| 06:03:45 | Russ | -P0 |
| 06:03:46 | TomW | Anyone built the blob lately. The one that comes with buildroot? |
| 06:03:51 | Russ | my uni blocks pings |
| 06:04:53 | jacques | TomW: not I - I dont mess with untested bootldr binaries - no JTAG |
| 06:04:58 | TomW | I built a blob, but it doesn't run. |
| 06:05:29 | Russ | TomW: 115200? |
| 06:05:42 | TomW | yup. |
| 06:06:06 | TomW | maybe it is stuck at 9600 for some reason.. |
| 06:06:17 | | rumour has it minicom is ok. In fact I did xmodem transfers from minicom. But after reseting, minicom receives dirty echo from serial port |
| 06:06:17 | Russ | minicom? |
| 06:06:50 | TomW | Minicom. I was able to put the blob-v0.4 back into Flash and it ran. |
| 06:07:04 | TomW | The config files look okay. |
| 06:09:55 | Russ | minicom should be at 115200 for latest blob |
| 06:10:16 | TomW | v2.0.4 is latest? |
| 06:11:13 | Russ | maybe release |
| 06:11:23 | Russ | I think its 2.0.5-pre something |
| 06:11:46 | Russ | 2.0.5-pre3 |
| 06:11:49 | jacques | heh once the tux has to start erasing flash, things get really slow |
| 06:12:32 | Russ | you need to reboot between flashing for some reason |
| 06:12:37 | Russ | its my bug I think |
| 06:13:01 | jacques | I'm just zmodeming my xserver binary over |
| 06:13:17 | jacques | first couple of times it's fine, but then it starts hanging while erasing |
| 06:13:50 | jacques | aha |
| 06:14:02 | jacques | it really is short circuiting the whole ts init... |
| 06:14:30 | Russ | oh, on the jffs2? |
| 06:14:37 | Russ | ya, it has to erase blocks |
| 06:14:45 | Russ | that has nothing to do with me |
| 06:15:00 | jacques | i know |
| 06:15:18 | jacques | just commenting that erasing takes a while |
| 06:17:19 | TomW | just putting the suspect blob code in now, I will try other baudrates. Linux doesn't come up, so I expect that blob crashes. |
| 06:20:04 | TomW | GNU Make version 3.79.1 |
| 06:21:02 | Russ | you can reblob |
| 06:21:52 | TomW | Yes, I take the binary from the TuxScreen site and use jflash to install it, then the board boots. |
| 06:22:11 | TomW | arm-linux-gcc v2.95.3, ? |
| 06:24:06 | TomW | Russ: are you using a precompiled blob, or one from the wget? |
| 06:24:27 | Russ | I compile my own from CVS |
| 06:26:13 | TomW | Ah, the instructions for the CVS version had me stumped, to many paths to enter. I didn't know if they wanted a path, or a fully qualified name (/usr/local/arm/2.95.3/bin OR /usr/local/arm/2.95.3/bin/arm-linux-gcc). |
| 06:27:12 | TomW | I guess that I will have to tackle the cvs version again and toss the buildroot version.. |
| 06:27:42 | jacques | i keep waiting for the jffs2-aware version to be blessed :-) |
| 06:59:35 | TomW | Hmmm... I now have a working blob-2.0.5-pre3, where is the download command! |
| 07:00:41 | | jg was last seen on #handhelds.org 23 days, 4 hours, 24 minutes and 40 seconds ago, saying: MonMotha: sorry... [Sat Mar 9 02:36:01 2002] |
| 07:00:41 | jacques | ibot: seen jg |
| 07:00:45 | | keithp was last seen on #handhelds.org 32 days, 3 hours, 11 minutes and 23 seconds ago, saying: is away: I'm not watching [Thu Feb 28 03:49:22 2002] |
| 07:00:45 | jacques | ibot: seen keithp |
| 07:16:38 | jacques | going back to older source tree - kdmouseinfo seems broken in recent cvs |
| 07:17:47 | TomW | What is the big deal abt the jffs2-aware version? |
| 07:18:01 | jacques | then we wont need the kernel partition |
| 07:18:15 | jacques | and can use modular kernels instead of a 900kB+ kernel |
| 07:18:29 | jacques | much more efficient use of onboard flash |
| 07:23:45 | BZFlag | TomW: you need a --enable=foo where foo is either xmodem or uuencode I think. |
| 07:24:01 | TomW | compresses the flash images, or just compresses it better (I assume it is not yet compressed). |
| 07:24:18 | TomW | BZFlag: Yup, got it, thanks. --enable-uucodec |
| 07:24:29 | BZFlag | TomW: xmodem is better. ;-) |
| 07:24:53 | BZFlag | TomW: flash is now: 128k blob, 1M-128k kernel, 3M rootfs. |
| 07:25:07 | BZFlag | rootfs is jffs2 compressed, read/write. |
| 07:25:46 | BZFlag | the plan is to make it: 128k blob+params, 4M-128k rootfs including kernel and modules. |
| 07:26:12 | BZFlag | so modules you need or don't need can be added or removed easily. |
| 07:27:01 | TomW | Hmmm.. I may try it in Xmodem, now that JTAG is working. BTW, it was the cable. I hauled out an 18 peice of 7 wire ribbon cable, terminated the wires with AMP terminals then inserted the wiring into the AMP housings.. Worked good. |
| 07:27:39 | BZFlag | jacques: current partitions could be resized, and modules will work at present, but I'd like to go to the final solution instead of another interim that will break compatibility. |
| 07:27:51 | jacques | BZFlag: agreed |
| 07:28:13 | TomW | <sigh> something wrong, the kernel won't boot after being flashed with the blob-2.0.5-pre3.. |
| 07:28:16 | BZFlag | TomW: excellent. feel free to leave comments on the wiki JTAG page. ;-) |
| 07:28:45 | BZFlag | here's my build steps for cvs blob: |
| 07:29:03 | TomW | send 'em over |
| 07:29:05 | BZFlag | tools/rebuild-gcc |
| 07:29:06 | BZFlag | tools/rebuild-gcc |
| 07:29:06 | BZFlag | CC=arm-linux-gcc |
| 07:29:06 | BZFlag | export CC |
| 07:29:06 | BZFlag | # --enable-blob-debug \ |
| 07:29:06 | BZFlag | ./configure \ |
| 07:29:09 | BZFlag | --with-linux-prefix=/home/timr/work/shannon/buildroot-tux/linux \ |
| 07:29:11 | BZFlag | --enable-maintainer-mode \ |
| 07:29:13 | BZFlag | --enable-clock-scaling \ |
| 07:29:15 | BZFlag | --enable-memtest \ |
| 07:29:19 | BZFlag | --enable-debug \ |
| 07:29:21 | BZFlag | --enable-lcd \ |
| 07:29:23 | BZFlag | --enable-jffs2 \ |
| 07:29:24 | BZFlag | --with-board=shannon \ |
| 07:29:26 | BZFlag | --enable-all-features \ |
| 07:29:28 | BZFlag | arm-linux |
| 07:29:30 | BZFlag | make |
| 07:29:39 | BZFlag | note that clock scaling and lcd don't work, so they get turned off anyway. |
| 07:31:51 | TomW | going back to old blob to ensure kernel & ramdisk are okay. Geez all this and still no money for doing embedded linux, soon I hope. :-( |
| 07:33:46 | TomW | BZFlag: I gave up on the other Tux board. The JTAG worked, but I couldn't put a good blob image into Flash, I could only get most of it in without corruption. |
| 07:34:43 | TomW | BZFlag: So, I tossed the board. I do have another Tux CPU that when you insert the SODIMM it won't boot. That one I will upgrade the DRAM in to 32Meg. |
| 07:37:25 | TomW | OHH! You don't use the arm-uclibc-gcc, you use the stock arm-linux-gcc. Maybe that is a problem. No reason I know of to use arm-uclibc-gcc as I don't need the environment to build under. |
| 07:38:35 | jacques | should build with either |
| 07:39:08 | TomW | k |
| 07:39:15 | jacques | well maybe not - I just cant think of a reason why uclibc-gcc would not work |
| 07:39:18 | BZFlag | yeah, should work with either, but as it does not use libs, arm-linux-gcc is probably easier. |
| 07:39:25 | jacques | agreed |
| 07:39:52 | TomW | have you noticed that 2.0.4 programs the flash faster than 2.0.5-pre3? |
| 07:43:47 | andersee | BZFlag: saw your comments yesterday re PATCH_DIR |
| 07:44:13 | andersee | BZFlag: Suggestions? |
| 07:44:25 | andersee | BZFlag: I'm not fixated on the idea per say... |
| 07:45:07 | jacques | hi andersee |
| 07:45:16 | andersee | BZFlag: I just wanted to recognize that patches (which tend to be stored locally in CVS) differ from source packages, which tend to be sucked down from elsewhere... |
| 07:45:22 | | bonjour, andersee |
| 07:45:22 | andersee | jacques: hi |
| 07:45:35 | andersee | jacques: making progress? |
| 07:46:54 | jacques | andersee: on xfree ts using tslib? I think so |
| 07:47:18 | andersee | really would like to run X on his phone... |
| 07:47:35 | BZFlag | andersee: I'd like to encourage the use of patches from elsewhere. ;-) |
| 07:48:15 | BZFlag | anyone try the inetd/telnetd in buildroot? |
| 07:48:28 | jacques | heh, it's kinda working |
| 07:48:31 | BZFlag | there's no script to start it yet, but the inetd.conf is there. |
| 07:48:43 | jacques | now we're cooking with gas |
| 07:49:16 | andersee | BZFlag: Take for example the busybox patch in CVS. I'm never going to put that into busybox itself... Where (other than the tuxscreen CVS) do you propose such things get pulled from? |
| 07:50:09 | BZFlag | I acknoledge that many times, perhaps even most times, the patches will be in cvs. |
| 07:50:20 | BZFlag | it's thinks like th multilayer kernel patch. |
| 07:51:04 | BZFlag | I might have used a uclinux telnetd patch is there was one rather than just another copy of the entire source. |
| 07:51:50 | BZFlag | any patches that can get pushed upstream should be. so using patches should not be elegant, lest we encourage them to persist. ;-) |
| 07:52:03 | andersee | Certainly for patches pulled from elsewhere (like for the kernel) we can treat those patches as simply source packages, per usual. Its the "local patches that live in CVS" I added PATCH_DIR to address |
| 07:52:11 | BZFlag | jacques: inetd/telnetd kinda working? or to what did you refer? |
| 07:53:00 | jacques | the tslib x stuff im working on |
| 07:53:12 | BZFlag | andersee: I don't see much point in treating them ay different. |
| 07:54:03 | BZFlag | eg: if russ (hi russ) want's to keep his kernel patches in buildroot cvs, cool. but imho it's even better for them to be web/ftp accessible. |
| 07:54:27 | BZFlag | other patches should be encouraged to exist outside of cvs if the maintainer so desires. |
| 07:54:34 | BZFlag | jacques: ahh. |
| 07:54:48 | andersee | BZFlag: They are very different. Stuff pulled from elsewhere is subject to deletion. Local patches are sacred and don't get the axe on a make clean... |
| 07:56:57 | TomW | jacques: put '/sbin/inetd &' in /linuxrc just before the 'exec /sbin/init', that will start inetd. |
| 08:01:15 | andersee | BZFlag: anyway, I'm not fixed on the idea. I was just trying to make a distinction between "local-and-sacred" vs "downloaded-and-volatile" |
| 08:13:33 | TomW | I am missing something with Xmodem upload: I start the upload 'xdownload kernel', then ^As, tag the file, then OKAY, it comes up with a prompt "press any key", I hit enter. |
| 08:14:02 | TomW | Then it prints: CCCCC.... |
| 08:14:02 | TomW | |
| 08:14:02 | TomW | |
| 08:14:02 | TomW | Loading kernel from flash .... done |
| 08:14:02 | TomW | *** invalid argument |
| 08:14:29 | TomW | checking /proc/interrupts shows that the serial port is not running as expected. |
| 08:19:32 | BZFlag | TomW: got lrzsz on you host system? |
| 08:21:17 | TomW | I do now.. |
| 08:23:03 | TomW | do I do something like 'cat tuxscreen-kernel | sx > /dev/ttyS0' ? |
| 08:28:49 | TomW | Ooookay, I see, minicom has no native xmodem protocol handler, lrzsz is the utility for minicom to do various protocols. :-/ |
| 08:30:26 | BZFlag | yepper. |
| 08:30:55 | TomW | hoo boy, brave new world. |
| 08:31:26 | TomW | five years of intensive linux learning and there is so much more to learn.. |
| 08:36:14 | andersee | TomW: I can sell you an Instant Linux Knowledge pill. Send $99.95 to me and I'll mail you a box of 'me. Buying them will make you much wiser. :-) |
| 08:37:15 | TomW | phzzzzt! And I have a bridge in Brooklyn ... |
| 08:37:37 | andersee | hehe |
| 08:38:34 | BZFlag | starts re-cooking ibot's logs |
| 08:38:40 | TomW | cheap! |
| 08:41:45 | andersee | http://www.google.com/technology/pigeonrank.html |
| 08:44:35 | TomW | andersee: Is that an April Fool joke? |
| 08:44:52 | andersee | I'm sure it is. But it was awfully funny. |
| 08:44:54 | TomW | naw! |
| 08:45:08 | andersee | Linked from the front page at google |
| 08:45:24 | TomW | At first it looks like an oversimplification of technology... |
| 08:46:31 | BZFlag | gotta love april fools. are the new rfc's out yet? |
| 08:46:46 | andersee | Havn't looked yet. |
| 08:48:01 | jacques | no |
| 08:48:02 | andersee | Nothing listed at http://www.rfc-editor.org/ |
| 08:48:13 | jacques | google used to be in same hosting facility as us |
| 08:48:19 | jacques | and that's how they do it |
| 08:48:25 | jacques | we moved partly beacuse of the smell |
| 08:48:44 | jacques | now we are a couple of cages away from ebay |
| 08:48:53 | jacques | and you dont wanna know what they use :-\ |
| 08:49:12 | andersee | 6 inches of pigeon poop on your switches? ;-) |
| 08:49:52 | jacques | yeah that sucks |
| 08:50:03 | TomW | does this make sense to anyone? |
| 08:50:06 | TomW | Fast Floating Point Emulator V0.9 (c) Peter Teichmann. |
| 08:50:06 | TomW | FAT: bogus logical sector size 381 |
| 08:50:06 | TomW | jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x000c3624: 0x1a1a instead |
| 08:50:07 | TomW | jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x000c3628: 0x1a1a instead |
| 08:50:07 | TomW | jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x000c362c: 0x1a1a instead |
| 08:51:04 | TomW | I had to do an 'erase kernel' & 'erase ramdisk' to get the blob-2.0.5 to program in a kernel & ramdisk. Do I need to 'erase parms' or something? |
| 08:51:28 | jacques | looks for april 1 jokes on cnn.com |
| 08:52:29 | andersee | jacques: End of the World -- reuters The world ended at 6pm yesterday. Reporters on the scene said .... |
| 08:53:26 | jacques | :-) |
| 08:59:53 | jacques | heh i just kicked the tuxscreen's ass in checkers |
| 09:00:34 | jacques | hmmm |
| 09:00:47 | jacques | just got an error trying to build buildroot with latest uclibc snap |
| 09:02:46 | andersee | oh? |
| 09:02:50 | andersee | details? |
| 09:03:57 | jacques | just a sec |
| 09:04:14 | andersee | is it with ftruncate64? |
| 09:04:25 | andersee | didn't test that one on arm |
| 09:05:10 | jacques | make[5]: Entering directory `/home/packages/tuxscreen/buildroot-tux/build/uClibc/libc/sysdeps/linux/arm' |
| 09:05:11 | jacques | arm-linux-gcc -Wall -Os -fstrict-aliasing -fno-builtin -nostdinc -pipe -nostdinc -I../../../../include -I/usr/local/arm/lib/gcc-lib/arm-linux/2.95.4/include -I. -D_LIBC -DNDEBUG -fPIC -D_LIBC_REENTRANT -c __longjmp.S -o __longjmp.o |
| 09:05:11 | jacques | /usr/local/arm/lib/gcc-lib/arm-linux/2.95.4/include/stddef.h: Assembler messages: |
| 09:05:11 | jacques | /usr/local/arm/lib/gcc-lib/arm-linux/2.95.4/include/stddef.h:170: Error: bad instruction `typedef unsigned int size_t' |
| 09:05:12 | jacques | ../../../../include/features.h:424: Error: bad instruction `extern int __path_search(char*tmpl,size_t tmpl_len,const char*dir,' |
| 09:05:15 | jacques | ../../../../include/features.h:425: Error: bad instruction `const char*pfx,int try_tmpdir)' |
| 09:05:17 | jacques | ../../../../include/features.h:426: Error: bad instruction `extern int __gen_tempname(char*__tmpl,int __kind)' |
| 09:05:20 | jacques | make[5]: *** [__longjmp.o] Error 1 |
| 09:05:48 | jacques | trying with 0.9.10 now |
| 09:06:32 | andersee | Bah. I knew I shouldn't have moved those prototypes. |
| 09:06:52 | andersee | K, I'll fix that in a minute... |
| 09:07:33 | jacques | 0.9.10 built fine |
| 09:09:20 | jacques | hmmm now busybox failed |
| 09:09:23 | jacques | that's strange |
| 09:09:30 | andersee | how? |
| 09:09:33 | TomW | Looks like the latest blob is buggy. You have to 'erase kernel' before it will program the kernel properly. |
| 09:09:39 | andersee | Do you have largefile enabled? |
| 09:10:00 | jacques | bunch of undefined references |
| 09:10:12 | jacques | only change I made to makefile was to use uclibc snap and I reverted that |
| 09:10:36 | jacques | all undefined are in nfsmount.o |
| 09:10:38 | andersee | I bet busybox needs a make clean |
| 09:11:13 | jacques | but it hadnt even been retrieved when I switched to working uclibc... |
| 09:11:23 | andersee | oh |
| 09:11:25 | TomW | goodnight all. |
| 09:11:25 | andersee | :( |
| 09:11:28 | andersee | night |
| 09:11:54 | jacques | nfsmount.o(.text+0xb4): undefined reference to `pmap_getmaps' |
| 09:11:55 | jacques | nfsmount.o: In function `nfsmount': |
| 09:11:55 | jacques | nfsmount.o(.text+0xd34): undefined reference to `clnt_perror' |
| 09:11:55 | jacques | nfsmount.o(.text+0xd84): undefined reference to `clnt_pcreateerror' |
| 09:11:55 | jacques | nfsmount.o(.text+0xec0): undefined reference to `clntudp_create' |
| 09:11:55 | jacques | nfsmount.o(.text+0xf10): undefined reference to `clnttcp_create' |
| 09:11:57 | jacques | nfsmount.o(.text+0xf30): undefined reference to `authunix_create_default' |
| 09:11:59 | jacques | nfsmount.o(.text+0x113c): undefined reference to `bindresvport' |
| 09:12:01 | jacques | nfsmount.o(.text+0x1190): undefined reference to `pmap_getport' |
| 09:12:03 | jacques | nfsmount.o(.text+0x12d0): undefined reference to `clnt_perror' |
| 09:12:05 | jacques | that's about half of them |
| 09:12:14 | andersee | Oh |
| 09:12:23 | jacques | look familiar at all? |
| 09:12:34 | andersee | busybox has nfs suppoprt enabled, but uClibc doesn't have RPC support enabled |
| 09:12:47 | jacques | heh |
| 09:12:59 | andersee | disable NFS in busybox, or enable RPC in uClibc |
| 09:13:06 | jacques | okay |
| 09:13:15 | jacques | what's best way to disable nfs ? |
| 09:13:37 | andersee | vi build/busybox*/Config.h |
| 09:13:38 | jacques | oh I se the patch in sources |
| 09:15:26 | jacques | it built now |
| 09:17:55 | jacques | ok now it's complaiining about not finding mkfs.jffs2 |
| 09:18:21 | jacques | do I need to install it separately now? |
| 09:18:44 | andersee | BZFlag was messing with that to try and get it to use the system one if available... |
| 09:19:03 | andersee | You probably need to edit make/jffs2root.mk |
| 09:19:23 | jacques | ok |
| 09:19:42 | BZFlag | jacques: it should dl and build mtd if it does not find it on your path. |
| 09:22:11 | jacques | BZFlag: but it does not |
| 09:22:30 | jacques | one thing I just changed was jffs2root was # out in Makefile as a target |
| 09:22:36 | jacques | but that did not fix the problem |
| 09:25:57 | | BZFlag: that doesn't look right |
| 09:25:57 | BZFlag | jacques: hmm... change the MKFS_JFFS2 define. |
| 09:28:06 | jacques | to what? |
| 09:29:15 | jacques | MKFS_JFFS2=$(shell which mkfs.jffs2 || echo $(MTD_DIR)/util/mkfs.jffs2) |
| 09:29:43 | jacques | there does not seem to be any provision for actually building the mtd tools |
| 09:30:08 | LoneTech | doesn't something depend on $MKFS_JFFS2 then? |
| 09:31:01 | jacques | heh I always get sidetracked |
| 09:31:18 | jacques | i really just wanted to build microwindows so I could see how the ts stuff is done |
| 09:36:32 | BZFlag | $(MTD_DIR)/util/mkfs.jffs2 |
| 09:37:29 | jacques | BZFlag: hmm? |
| 09:38:17 | LoneTech | there's TS stuff in picogui too |
| 09:39:28 | BZFlag | set: MKFS_JFFS2=$(MTD_DIR)/util/mkfs.jffs2 |
| 09:39:38 | jacques | LoneTech: but does it work? I heard it wasnt working |
| 09:39:55 | scanline | is away: standing in the wet room |
| 09:40:27 | LoneTech | it has worked.. I haven't tried it on the tuxscreen yet |
| 09:40:48 | LoneTech | I wrote the shareable calibration stuff and the palmax driver |
| 09:40:53 | jacques | LoneTech: i meant specifically on the tuxscreen with recent kernels |
| 09:41:05 | LoneTech | I believe there was some confusion with two tuxscreen drivers because of kernel changes. |
| 09:41:31 | jacques | they changed the kernel driver around 2.4.10 - and X still does not work |
| 09:43:50 | LoneTech | ic |
| 09:44:10 | LoneTech | well, once I start thinking straight I'll fix picogui if it's broken. |
| 09:44:22 | LoneTech | unless you have a problem with that |
| 09:44:46 | LoneTech | (haven't quite woken up yet) |
| 09:45:17 | jacques | i have no problem with anyone fixing anything |
| 09:45:23 | jacques | esp ts drivers |
| 09:45:28 | jacques | i am trying to get x working |
| 09:45:32 | jacques | with limited success |
| 09:45:47 | jacques | tslib is bahaving strangely |
| 09:46:39 | jacques | and I see things like this in the microwin driver code: |
| 09:46:41 | jacques | /* |
| 09:46:41 | jacques | * kernel driver bug: select returns read available, |
| 09:46:41 | jacques | * but read returns -1 |
| 09:47:58 | LoneTech | read should only return -1 on some form of exceptional status |
| 09:48:10 | LoneTech | non-blocking EAGAIN is one though. |
| 09:48:32 | LoneTech | I'll look at it |
| 09:48:41 | LoneTech | but, as always, no guarantees |
| 09:49:58 | jacques | no problem |
| 09:50:13 | jacques | I am just trying to figure out some of my x problems by looking at working code |
| 09:55:36 | LoneTech | good luck |
| 09:57:06 | jacques | thanks |
| 10:02:44 | scanline | is back (gone 00:22:56) |
| 10:14:22 | jacques | hmm i dont know why but the tslib version is very slow compared to the hack on the wiki |
| 14:00:11 | scanline | is away: pseudo-foodlike-thingies |
| 14:18:26 | scanline | is back (gone 00:18:15) |
| 16:27:41 | prpplague | morning all |
| 16:36:07 | CosmicPenguin | prpplague: morning |
| 16:58:15 | CosmicPenguin | Ahhh - April Fools jokes... what fun.. :) |
| 17:23:53 | scanline | CosmicPenguin: what was april fools, the morning? |
| 17:32:42 | scanline | ah... slashdot has its fair share of lame april fools jokes |
| 17:33:01 | scanline | and a good one or two |
| 17:40:58 | scanline | laughs at today's User Friendly |
| 17:46:04 | scanline | is away: class |
| 18:08:45 | CosmicPenguin | scanline: yeah - and the yearly spoofed e-mail from Linus Torvalds on the kernel mailing list was pretty funny too |
| 18:09:22 | CosmicPenguin | The best is the Google one: http://www.google.com/technology/pigeonrank.html |
| 19:44:00 | TomW | farts |
| 21:05:23 | sjhill | andersee: hey Erik |
| 21:05:35 | andersee | sjhill: howdy |
| 21:05:40 | andersee | sjhill: how goes the war? |
| 21:05:44 | sjhill | andersee: great |
| 21:05:49 | andersee | sjhill: :-) |
| 21:05:58 | sjhill | andersee: i almost have the dynamic linker bootstrapped |
| 21:06:12 | andersee | sjhill: Wohoo! mips users rejoice! |
| 21:06:17 | sjhill | heh |
| 21:13:14 | andersee | sjhill: Out of curiosity, is somebody paying you to work on the mips ld.so, or are you doing it for fun/personal enrichment? |
| 21:13:27 | andersee | sjhill: Just curious... |
| 21:13:53 | sjhill | andersee: i'm being paid.....i mean what sick and psychopathic person would want to write a dynamic linker/loader for fun? :) |
| 21:14:26 | andersee | hehe. |
| 21:14:36 | andersee | feels sick |
| 21:14:40 | sjhill | lol |
| 21:14:52 | sjhill | andersee: the best part is, i get to release everything |
| 21:15:10 | andersee | Though in truth, I was paid while working on the ldso arm port |
| 21:15:19 | andersee | sjhill: excellent. |
| 21:25:30 | andersee | howdy BZFlag |
| 21:25:47 | andersee | How're things at lineo today? |
| 21:33:59 | BZFlag | quiet |
| 21:47:12 | prpplague | BZFlag: have your choice of employee lockers these days? |
| 21:54:55 | CosmicPenguin | methinks that Slashdot is wearing itself a bit thin... the jokes are getting old |
| 21:57:47 | prpplague | CosmicPenguin: ya recken the story about will wheaton getting a part on enterprises is true or is that another A1 joke? |
| 22:11:09 | prpplague | mallum: dude! |
| 22:11:28 | prpplague | mallum: how goes it |
| 22:11:40 | prpplague | mallum: whats the latest on matchbox? |
| 22:13:01 | CosmicPenguin | prpplague: thats gotta be a A1 joke |
| 22:18:12 | mallum | prpplague: hi :) |
| 22:18:37 | prpplague | CosmicPenguin: i alway thought they would do that with voyager |
| 22:18:39 | mallum | prpplague: just figured out how to guarentee ordering of X app startup |
| 22:18:46 | prpplague | mallum: nice |
| 22:19:00 | prpplague | mallum: threaten to beat them? |
| 22:19:17 | mallum | prpplague: I found a little tool called 'xtoolwait' |
| 22:19:29 | prpplague | mallum: do tell |
| 22:19:45 | mallum | prpplague: which basically lauches an app, waits on a createwindow event, then a property notify |
| 22:19:51 | mallum | prpplague: then returns |
| 22:20:04 | mallum | prpplague: this is cool cos now I can orders apps in matchbox's dock |
| 22:20:14 | mallum | prpplague: and not worry about varying startup times |
| 22:20:28 | mallum | prpplague: or lots of sleep 1 & minitime & etc |
| 22:20:31 | prpplague | mallum: that is nice, i need that as well, where didja get it? |
| 22:20:47 | mallum | prpplague: its in debian |
| 22:21:26 | mallum | prpplague: source is at http://www.hacom.nl/~richard/software/xtoolwait.html |
| 22:22:21 | mallum | prpplague: it also means I can hack monoluanch so you dont have to supply the window name ( I think ) |
| 22:22:34 | prpplague | bookmarks |
| 22:24:12 | mallum | I build gnome2 today |
| 22:24:18 | mallum | it took nearly 5 hours |
| 22:24:29 | mallum | and looks nice, but not really to my taste |
| 22:26:45 | prpplague | mallum: ya, i use icewm mostly, but gnome sometimes |
| 22:27:45 | mallum | prpplague: yeah I just use a hacked blackbox |
| 22:28:17 | mallum | prpplague: now I have this new machine with some spare cpu, I though a 'desktop' would be nice |
| 22:28:44 | mallum | prpplague: I dont think blackbox supports desktops though - not gnome ones anyway |
| 23:44:52 | prpplague | sjm: morning |
| 23:45:02 | prpplague | sjm: hows life down under today? |
| 23:45:17 | sjm | prpplague: Howdy, nice and relaxed after four days off :) |
| 23:51:27 | prpplague | sjm: done any good arm hack lately? |
| 23:51:47 | prpplague | sjm: i got my maxim samples in but had no time to play :( |
| 23:53:35 | sjm | prpplague: Not hacking on arm atm, doing 68HC705 stuff for work, and x86 serial port stuff at home. |
| 23:54:31 | sjm | prpplague: I'm writing a screen capture program for my Fluke ScopeMeter that works under linux. I hate having to fire up MS just to get a screendump. |
| 23:55:20 | sjm | prpplague: I had to debug the MS-fluke serial protocol which took a few days... |